562 - Neonatal Morbidity and Mortality in a Contemporary Cohort of Premature Infants at High Risk for Infection

Poster Number: 562
Publication Number: 562.336

Danielle Browning, UPMC Childrens Hospital of Pittsburgh, Pittsburgh, PA, United States; Alexandria Sasaki, University of Pittsburgh School of Medicine, Pittsburgh, PA, United States; Christine E. Bishop, UPMC Childrens Hospital of Pittsburgh, Pittsburgh, PA, United States; Toby D. Yanowitz, University of Pittsburgh School of Medicine, Pittsburgh, PA, United States; Christina Megli, University of Pittsburgh School of Medicine, Pittsburgh, PA, United States

Background: Antibiotic administration to pregnant women in preterm premature rupture of membranes (PPROM) is associated with many physiologic and microbiologic changes in both the women and infants. PPROM accounts for one third of all preterm deliveries. The study of neonatal outcomes in PPROM, including early onset sepsis (EOS), late onset sepsis (LOS), necrotizing enterocolitis (NEC) and neonatal death has not been reported in a large population in 20 years. In the last 20 years, we have seen an increase in survival to discharge of periviable neonates and changes in the rates of EOS, LOS and NEC (≥2A).

Objective: To define the rates of neonatal complications (Table 1) in a contemporary cohort of neonates born to pregnant women with PPROM. In addition, to define these outcomes dependent on gestational age (GA) at PPROM.

Design/Methods:
Inclusion criteria for this retrospective cohort were ICD-9 (658.10) and ICD-10 (O42.913) for PPROM between July 2015 and January 2018 and gestational age (GA) at delivery 23-34 weeks. Exclusion criteria were erroneous coding, GA at delivery < 23 weeks or >34 weeks, PPROM >33 6/7, and congenital anomalies. A detailed chart review for neonatal characteristics was performed. Patients were stratified by GA at PPROM to compare groups: < 23 weeks, 23 0/7-24 6/7, 25 0/7-27 6/7, 28 0/7-30 6/7, and 31-33 6/7. Chi square test for trend and chi square on contingency tables were used to compare dichotomous variables. ANOVA was used to compare continuous variables. All statistical analysis was performed using graph pad prism. Outcomes compared include EOS, LOS, NEC ≥2A, IVH grade ≥3, composite primary outcome and total composite adverse outcome.
 

Results: 246 infants met inclusion criteria. The mean GA at birth and latency was significantly different between groups (Figure 1). The rate of death was statistically significant for the previable group. Other outcomes including NEC, LOS, and EOS were not significant by gestational age at PPROM (Table 1). In addition, the rates for composite primary outcomes, BPD, and total composite adverse outcomes were all significant (p < 0.0001).Conclusion(s): Composite adverse outcomes are greater than expected in this population at 22.8%, for which BPD and death are the leading contributors. A high rate of infants with previable and periviable PPROM died despite achieving a long latency. In addition, the duration of latency amongst our cohort is much longer than previously reported in earlier studies (Figure 2). Further studies looking at determinants of latency and neonatal outcomes are indicated.
CV Danielle BrowningCV- Danielle Naomi Simpson Browning.pdf
Figure 1. Kaplan-Meier survival curve of pregnancy latency days after PPROM in < 23 weeks, 23w0d-24w6d, 25w0d-27w6d, 28w0d-30w6d and 31w0d-33w6d.